Relatório Comparativo: Modelos de Classificação

Modelo 1: Logistic Regression (Baseline) Voltar ao topo

Médias das Métricas

• Accuracy: 0.639
• Precision: 0.167
• Recall: 0.507
• F1-Score: 0.226
• ROC AUC: 0.579

Curvas ROC

dataset_1000_hypothyroid.csv
AUC = 0.541

dataset_1002_ipums_la_98-small.csv
AUC = 0.490

dataset_1004_synthetic_control.csv
AUC = 0.793

dataset_1013_analcatdata_challenger.csv
AUC = 0.500

dataset_1014_analcatdata_dmft.csv
AUC = 0.479

dataset_1016_vowel.csv
AUC = 0.654

dataset_1018_ipums_la_99-small.csv
AUC = 0.449

dataset_1020_mfeat-karhunen.csv
AUC = 0.707

dataset_1021_page-blocks.csv
AUC = 0.703

dataset_1022_mfeat-pixel.csv
AUC = 0.601

dataset_1023_soybean.csv
AUC = 0.476

dataset_1039_hiva_agnostic.csv
AUC = 0.539

dataset_1045_kc1-top5.csv
AUC = 0.833

dataset_1049_pc4.csv
AUC = 0.483

dataset_1050_pc3.csv
AUC = 0.526

dataset_1056_mc1.csv
AUC = 0.385

dataset_1059_ar1.csv
AUC = 0.309

dataset_1061_ar4.csv
AUC = 0.648

dataset_1064_ar6.csv
AUC = 0.681

dataset_1065_kc3.csv
AUC = 0.589

dataset_311_oil_spill.csv
AUC = 0.601

dataset_312_scene.csv
AUC = 0.555

dataset_316_yeast_ml8.csv
AUC = 0.625

dataset_38_sick.csv
AUC = 0.466

dataset_450_analcatdata_lawsuit.csv
AUC = 0.770

dataset_463_backache.csv
AUC = 0.687

dataset_757_meta.csv
AUC = 0.599

dataset_764_analcatdata_apnea3.csv
AUC = 0.467

dataset_765_analcatdata_apnea2.csv
AUC = 0.511

dataset_767_analcatdata_apnea1.csv
AUC = 0.340

dataset_865_analcatdata_neavote.csv
AUC = 0.696

dataset_867_visualizing_livestock.csv
AUC = 0.605

dataset_875_analcatdata_chlamydia.csv
AUC = 0.542

dataset_940_water-treatment.csv
AUC = 0.569

dataset_947_arsenic-male-bladder.csv
AUC = 0.500

dataset_949_arsenic-female-bladder.csv
AUC = 0.479

dataset_950_arsenic-female-lung.csv
AUC = 0.833

dataset_951_arsenic-male-lung.csv
AUC = 0.235

dataset_954_spectrometer.csv
AUC = 0.703

dataset_958_segment.csv
AUC = 0.675

dataset_962_mfeat-morphological.csv
AUC = 0.412

dataset_966_analcatdata_halloffame.csv
AUC = 0.610

dataset_968_analcatdata_birthday.csv
AUC = 0.532

dataset_971_mfeat-fourier.csv
AUC = 0.659

dataset_976_JapaneseVowels.csv
AUC = 0.663

dataset_978_mfeat-factors.csv
AUC = 0.753

dataset_980_optdigits.csv
AUC = 0.670

dataset_984_analcatdata_draft.csv
AUC = 0.570

dataset_987_collins.csv
AUC = 0.459

dataset_995_mfeat-zernike.csv
AUC = 0.797

Modelo 2: Logistic Regression (After Changes) 🔝 Voltar ao topo

Médias das Métricas

• Accuracy: 0.935
• Precision: 0.707
• Recall: 0.429
• F1-Score: 0.490
• ROC AUC: 0.709

Curvas ROC

dataset_1000_hypothyroid.csv
AUC = 0.621

dataset_1002_ipums_la_98-small.csv
AUC = 0.513

dataset_1004_synthetic_control.csv
AUC = 1.000

dataset_1013_analcatdata_challenger.csv
AUC = 0.500

dataset_1014_analcatdata_dmft.csv
AUC = 0.500

dataset_1016_vowel.csv
AUC = 0.704

dataset_1018_ipums_la_99-small.csv
AUC = 0.520

dataset_1020_mfeat-karhunen.csv
AUC = 0.982

dataset_1021_page-blocks.csv
AUC = 0.715

dataset_1022_mfeat-pixel.csv
AUC = 0.994

dataset_1023_soybean.csv
AUC = 0.858

dataset_1039_hiva_agnostic.csv
AUC = 0.607

dataset_1045_kc1-top5.csv
AUC = 0.738

dataset_1049_pc4.csv
AUC = 0.608

dataset_1050_pc3.csv
AUC = 0.517

dataset_1056_mc1.csv
AUC = 0.500

dataset_1059_ar1.csv
AUC = 0.500

dataset_1061_ar4.csv
AUC = 0.713

dataset_1064_ar6.csv
AUC = 0.600

dataset_1065_kc3.csv
AUC = 0.538

dataset_311_oil_spill.csv
AUC = 0.583

dataset_312_scene.csv
AUC = 0.902

dataset_316_yeast_ml8.csv
AUC = 0.500

dataset_38_sick.csv
AUC = 0.521

dataset_450_analcatdata_lawsuit.csv
AUC = 0.667

dataset_463_backache.csv
AUC = 0.632

dataset_757_meta.csv
AUC = 0.490

dataset_764_analcatdata_apnea3.csv
AUC = 0.683

dataset_765_analcatdata_apnea2.csv
AUC = 0.737

dataset_767_analcatdata_apnea1.csv
AUC = 0.710

dataset_865_analcatdata_neavote.csv
AUC = 0.500

dataset_867_visualizing_livestock.csv
AUC = 0.754

dataset_875_analcatdata_chlamydia.csv
AUC = 0.896

dataset_940_water-treatment.csv
AUC = 0.604

dataset_947_arsenic-male-bladder.csv
AUC = 0.714

dataset_949_arsenic-female-bladder.csv
AUC = 0.559

dataset_950_arsenic-female-lung.csv
AUC = 0.750

dataset_951_arsenic-male-lung.csv
AUC = 0.875

dataset_954_spectrometer.csv
AUC = 0.824

dataset_958_segment.csv
AUC = 0.724

dataset_962_mfeat-morphological.csv
AUC = 0.992

dataset_966_analcatdata_halloffame.csv
AUC = 0.820

dataset_968_analcatdata_birthday.csv
AUC = 0.646

dataset_971_mfeat-fourier.csv
AUC = 0.998

dataset_976_JapaneseVowels.csv
AUC = 0.885

dataset_978_mfeat-factors.csv
AUC = 0.993

dataset_980_optdigits.csv
AUC = 0.939

dataset_984_analcatdata_draft.csv
AUC = 0.500

dataset_987_collins.csv
AUC = 0.839

dataset_995_mfeat-zernike.csv
AUC = 0.990

Modelo 3: Logistic Regression (With SMOTE) 🔝 Voltar ao topo

Médias das Métricas

• Accuracy: 0.785
• Precision: 0.359
• Recall: 0.845
• F1-Score: 0.477
• ROC AUC: 0.881

Curvas ROC

dataset_1000_hypothyroid.csv
AUC = 0.905

dataset_1002_ipums_la_98-small.csv
AUC = 0.840

dataset_1004_synthetic_control.csv
AUC = 1.000

dataset_1013_analcatdata_challenger.csv
AUC = 0.667

dataset_1014_analcatdata_dmft.csv
AUC = 0.506

dataset_1016_vowel.csv
AUC = 0.976

dataset_1018_ipums_la_99-small.csv
AUC = 0.881

dataset_1020_mfeat-karhunen.csv
AUC = 1.000

dataset_1021_page-blocks.csv
AUC = 0.939

dataset_1022_mfeat-pixel.csv
AUC = 1.000

dataset_1023_soybean.csv
AUC = 0.980

dataset_1039_hiva_agnostic.csv
AUC = 0.770

dataset_1045_kc1-top5.csv
AUC = 0.929

dataset_1049_pc4.csv
AUC = 0.859

dataset_1050_pc3.csv
AUC = 0.825

dataset_1056_mc1.csv
AUC = 0.927

dataset_1059_ar1.csv
AUC = 0.843

dataset_1061_ar4.csv
AUC = 0.790

dataset_1064_ar6.csv
AUC = 0.585

dataset_1065_kc3.csv
AUC = 0.845

dataset_311_oil_spill.csv
AUC = 0.902

dataset_312_scene.csv
AUC = 0.966

dataset_316_yeast_ml8.csv
AUC = 0.910

dataset_38_sick.csv
AUC = 0.928

dataset_450_analcatdata_lawsuit.csv
AUC = 1.000

dataset_463_backache.csv
AUC = 0.726

dataset_757_meta.csv
AUC = 0.746

dataset_764_analcatdata_apnea3.csv
AUC = 0.936

dataset_765_analcatdata_apnea2.csv
AUC = 0.878

dataset_767_analcatdata_apnea1.csv
AUC = 0.920

dataset_865_analcatdata_neavote.csv
AUC = 0.696

dataset_867_visualizing_livestock.csv
AUC = 0.790

dataset_875_analcatdata_chlamydia.csv
AUC = 0.965

dataset_940_water-treatment.csv
AUC = 0.819

dataset_947_arsenic-male-bladder.csv
AUC = 0.841

dataset_949_arsenic-female-bladder.csv
AUC = 0.814

dataset_950_arsenic-female-lung.csv
AUC = 0.832

dataset_951_arsenic-male-lung.csv
AUC = 1.000

dataset_954_spectrometer.csv
AUC = 0.983

dataset_958_segment.csv
AUC = 0.989

dataset_962_mfeat-morphological.csv
AUC = 0.989

dataset_966_analcatdata_halloffame.csv
AUC = 0.966

dataset_968_analcatdata_birthday.csv
AUC = 0.949

dataset_971_mfeat-fourier.csv
AUC = 1.000

dataset_976_JapaneseVowels.csv
AUC = 0.986

dataset_978_mfeat-factors.csv
AUC = 1.000

dataset_980_optdigits.csv
AUC = 0.990

dataset_984_analcatdata_draft.csv
AUC = 0.491

dataset_987_collins.csv
AUC = 0.952

dataset_995_mfeat-zernike.csv
AUC = 1.000

🔍 Comparação por Dataset